PDF OCR 辨識

將掃描的 PDF 頁面轉換為可編輯、可搜尋的文字 — 從任何掃描文件、純影像 PDF 或頁面照片中擷取文字。整個過程在您的瀏覽器中本地執行。

OCR 準確率取決於影像品質。為獲得最佳效果,請使用 300 DPI 或更高的清晰掃描檔。手寫文字和藝術字體的辨識準確率可能較低。

將掃描的 PDF 拖到此處,或點擊選擇檔案

選擇 PDF 檔案

最大檔案大小:128 MB

所有處理在您的瀏覽器中本地完成。檔案永遠不會上傳。

掃描書 PDF OCR — 讓影像型 PDF 可搜尋

Need to search inside a scanned book, textbook, or printed document? Our browser-side OCR turns image-only PDFs into searchable PDFs with an invisible text layer — no upload, no account, works on multi-hundred-page files. Perfect for textbook study, legal research, and citation work.

100% FreeNo UploadNo Sign-upSearchable PDF Output

使用步驟

Upload Scanned PDF: Drop the image-only PDF (from phone scans, flatbed scanners, or library digital copies) into the tool. PDFs up to 500 MB are supported by splitting the OCR into chunks internally.
Choose OCR Language: Select the original language of the book. For mixed-language academic papers (e.g. English with some Latin/French quotes), use the 'Auto-detect' mode. Multi-language packs are supported.
Run OCR: Click 'Start OCR'. Tesseract.js processes each page in your browser via WebAssembly. A 200-page book typically completes in 5-10 minutes on a modern device.
Download Searchable PDF: Save the OCR'd PDF. It now contains a selectable text layer while preserving the original scanned image. You can search (Ctrl+F) and copy text from any page. The file size grows ~20-40% due to the text layer.

為何選擇此工具

100% Local Processing: Files are processed entirely in your browser using JavaScript — never uploaded to any server.
No Limits: No file count or file size restrictions. Process as many files as your device can handle.
No Sign-up: Free forever, no account needed, no email required. Open the page and start.
Private by Design: Nothing is sent to any server. Close the tab and your files are gone forever.

常見問題

Can I OCR a scanned PDF?

Yes. Our PDF OCR tool converts scanned (image-only) PDFs into searchable PDFs with selectable text. Drop your file, click 'Start OCR', and within minutes the result contains both the original image and an invisible text layer for search/copy.

Will the OCR lose the original image quality?

No — our tool keeps the original scanned pages as-is and overlays the recognized text on top. You see the same images as before, but now Ctrl+F finds words and you can copy any text. File size grows modestly (20-40%) due to the added text layer.

What languages does the PDF OCR support?

Our OCR supports 100+ languages including English, Spanish, French, German, Italian, Portuguese, Chinese (Simplified & Traditional), Japanese, Korean, Arabic, Hebrew, Russian, Hindi, and many more. Multi-language pages are handled via auto-detection.

How accurate is OCR on scanned books?

Accuracy depends on scan quality. Clean scans at 300 DPI typically achieve 95-98% accuracy for printed text in Latin scripts. Cursive, low-contrast scans, or stylized fonts drop to 80-90%. The output text is plain — no formatting, italics, or bold are preserved.

How long does OCR take on a 200-page book?

On a modern laptop, expect 5-10 minutes for a 200-page book. Older phones may take 15-20 minutes. The tool runs entirely in your browser — your device does all the work; nothing is uploaded.

Will my scanned book be uploaded?

No. OCR processing happens 100% locally in your browser via Tesseract.js WebAssembly. Your PDF and OCR results never leave your device. This is critical for sensitive documents: medical records, legal files, proprietary research, personal correspondence.

Can I search the resulting OCR PDF?

Yes — that's the main benefit. Open the output PDF in any reader (Adobe, Preview, Chrome, browser PDF viewer, mobile apps) and Ctrl+F / Cmd+F searches the recognized text layer. The original image stays intact underneath.